Pinned Tweet
Introducing Unsloth Desktop š¦„
The first desktop app to run and train models locally.
⢠Open-source. Runs on Mac, Windows and Linux
⢠Supports MLX, diffusion image/video, audio, GGUF
⢠Connect Claude Code and Codex to local LLMs
⢠50% more accurate, self-healing tool calls + sandboxed code exec
⢠Works for CPU + multiGPU setups - NVIDIA, AMD, Intel, Mac
⢠Train models 2à faster with 70% less VRAM
⢠Private web search, deep research, RAG, MCP and exports (NVFP4, GGUF)
⢠Use Unslothās OpenAI-compatible API and cloud models
⢠Securely deploy LLMs remotely and access anywhere
Unsloth Desktop is now available on unsloth.ai and GitHub.
GitHub: github.com/unslothai/unsloth
Blog and Guide: unsloth.ai/docs/desktop
Qwen announces smaller Qwen4 models will be released soon! š„
Qwen4-27B will run on 16GB RAM setups.
š¢ Meet Qwen4-Max, your new AI teammate for work, creativity, and discovery!
Qwen4-Max is built on a new MoE architecture with native vision and a 1M-token context window.
Highlights:
- Outperforms Qwen3.8-Max with up to ~50% fewer tokens, while average cost per task drops over 60% in our internal evaluation.
- Advances real-world productivity through greater proficiency with professional tools, stronger long-horizon execution, and higher-quality deliverables.
- Handles real tasks, including 3D modeling, game development, manufacturing, finance, law, healthcare, and frontier scientific discovery.
- $1.5/M tokens for input, $4.5/MĀ tokens for output, and $0.075/M tokens forĀ implicitĀ cachingĀ ā allĀ lower than Qwen3.8-Max, withĀ implicit cachingĀ downĀ by 70%.
The API is live on QwenCloud. You can also try Qwen4-Max on Qwen Studio, QwenWork and Qoder.
Start building today! š
- Blog: qwen.ai/blog?id=qwen4-max
- API: qwencloud.com/models/qwen4-mā¦
- Token Plan: qwencloud.com/pricing/token-ā¦
More to come! The open weights of smaller Qwen4 models will be released. Stay tuned. š
Qwen4 GGUFs will use a new version of Unsloth Dynamic 3.1 for the best performance on Terminal Bench, DeepSWE & more agentic benchmarks.
Today, we're releasing the highest accuracy Qwen3.8-27B GGUFs.
Our new Dynamic V3.1 GGUFs lead by 5ā10% on average across Terminal Bench, DeepSWE & more. Weāre working on even better version for Qwen4
Run on 12GB RAM.
GGUF: huggingface.co/unsloth/Qwen3ā¦
Blog: unsloth.ai/docs/models/qwen3ā¦
Today, we're releasing the highest accuracy Qwen3.8-27B GGUFs.
Our new Dynamic V3.1 GGUFs lead by 5ā10% on average across Terminal Bench, DeepSWE & more. Weāre working on even better version for Qwen4
Run on 12GB RAM.
GGUF: huggingface.co/unsloth/Qwen3ā¦
Blog: unsloth.ai/docs/models/qwen3ā¦
Qwen3.8-Flash-Next GGUFs now run faster with less thinking and less token use.
Unsloth Dynamic V3.1 GGUFs bring thinking & token use to match 95% of full precision Qwen3.8-Flash.
Results: 50% fewer tokens + 15% higher accuracy vs same size quants.
GGUF: huggingface.co/unsloth/Qwen3ā¦
Weāre releasing new Qwen3.8-Flash-Next GGUFs with 15% higher accuracy.
Unsloth Dynamic V3.1 outperforms others by >15% on Terminal Bench, DeepSWE & more benchmarks (256K context).
Run on 32GB+ RAM setups.
GGUF: huggingface.co/unsloth/Qwen3ā¦
Blog: unsloth.ai/docs/models/qwen3ā¦
Weāre releasing new Qwen3.8-Flash-Next GGUFs with 15% higher accuracy.
Unsloth Dynamic V3.1 outperforms others by >15% on Terminal Bench, DeepSWE & more benchmarks (256K context).
Run on 32GB+ RAM setups.
GGUF: huggingface.co/unsloth/Qwen3ā¦
Blog: unsloth.ai/docs/models/qwen3ā¦
We benchmarked DeepSeek-V4.1-Flash-1017 GGUFs on Terminal Bench, DeepSWE & more on 512K context. Unsloth ranks highest overall.
1-bit Unsloth GGUF scores 68.1 on DeepSWE, 72.2 on Terminal-Bench, 52.7 on AutomationBench. >10% ahead of other quants.
GGUF: huggingface.co/unsloth/DeepSā¦
Weāre releasing new DeepSeek-V4.1-Flash-1017 GGUFs, ranking highest across benchmarks.
Dynamic V3.1 leads across nearly all benchmarks: GPQA Diamond, AIME25, LiveCodeBench v6, MMLU-Pro, Aider & more.
Run on 64GB+ RAM.
GGUF: huggingface.co/unsloth/DeepSā¦
Blog: unsloth.ai/docs/models/deepsā¦
Weāre releasing new DeepSeek-V4.1-Flash-1017 GGUFs, ranking highest across benchmarks.
Dynamic V3.1 leads across nearly all benchmarks: GPQA Diamond, AIME25, LiveCodeBench v6, MMLU-Pro, Aider & more.
Run on 64GB+ RAM.
GGUF: huggingface.co/unsloth/DeepSā¦
Blog: unsloth.ai/docs/models/deepsā¦
š DeepSeek-V4.1-Flash-1017 is here!
Stronger benchmark performance, faster task completion, and shorter outputs than the previous versionāa Pareto improvement in performance and cost. Now live on our website, app, and API.
š ļø Optimized for @DeepSeekHarness, Tencent CodeBuddy, WorkBuddy, and more. Upgrade to DeepSeek Harness v0.2.1-rc.1 or later for the best experience with the new model.
1/4
Mistral Le Chonk can now run locally!
The 1 trillion parameter model runs on a Mac Studio with our 1-bit GGUF which retains 91% top-1 accuracy.
Le Chonk is the strongest Western open model to date.
GGUF: huggingface.co/unsloth/Mistrā¦
Guide: unsloth.ai/docs/models/mistrā¦
Today, we release the weights of a new and improved Le Chonk, advancing the open frontier for agentic coding and cybersecurity.
š» Built to build: Le Chonk ranks #1 on @cognitionās FrontierCode among open models, and sets a new benchmark for Pareto efficiency on @artificialanalysis.
š” Built to defend: State-of-the-art in cyberdefense, put through the paces by the world's leading cyber vendors, including @CrowdStrike, @cloudflare, @rubrik, @zscaler, and others.
Built for sovereignty: Make it yours. Download the weights on @huggingface , customize it with Mistral Forge, and deploy it wherever you need it.
Read more from our team and customers š:
Your local LLM can now take control of a browser and complete tasks autonomously via Unsloth!
Qwen3.8-27B successfully controlled the browser to complete tasks like chess, painting, with no human intervention.
GitHub: github.com/unslothai/unsloth
Guide: unsloth.ai/docs/basics/browsā¦
Unsloth AI retweeted
MiniMax M3.1 Flash meets the @UnslothAI community. šš¦„āØ
Unslothās Dynamic GGUFs bring our 428B MoE model to a 143.9GB 2-bit footprint, with 91.5% top-1 token agreement in their evaluation across seven domains.
That means a lower memory barrier to self-hosting, deployment through Unsloth Desktop and llama.cpp, and an OpenAI-compatible API for existing apps and agents.
With 78K+ GitHub stars and over 500M model downloads on Hugging Face, Unsloth gives M3.1 Flash a path into an established community of developers running, evaluating and building with open models.
Open weights. Local control. More ways to build.
MiniMax-M3.1-Flash can now run locally!
We made 2-bit GGUF retain 91.5% top 1 accuracy while being 45% smaller. Run on 150GB RAM.
MiniMax-M3.1-Flash is the strongest open model for its size & among the best overall.
GGUF: huggingface.co/unsloth/MiniMā¦
Guide: unsloth.ai/docs/models/minimā¦
Unsloth AI retweeted
Getting started with @UnslothAI on RTX Spark? Just run the .exe.
@danielhanchen and @NaderLikeLadder got hands-on with NVIDIA RTX Spark to talk about running and training AI models locally, exploring reinforcement learning, and making it easier to start building with Unsloth.
MiniMax-M3.1-Flash can now run locally!
We made 2-bit GGUF retain 91.5% top 1 accuracy while being 45% smaller. Run on 150GB RAM.
MiniMax-M3.1-Flash is the strongest open model for its size & among the best overall.
GGUF: huggingface.co/unsloth/MiniMā¦
Guide: unsloth.ai/docs/models/minimā¦
MiniMax M3.1 Flash: Space Performance + Bunny Speed
šš° Yes, Space Bunny - which has topped global leaderboards with 200T+ tokens - was an earlier version of M3.1 Flash.
š°Redefining the performance economics of agentic AI. M3.1 Flash achieves frontier-level intelligence across key tasks such as Cowork (Automation-Bench: 62%; ALE: 30%), Coding (DeepSWE: 74%) and Visual Reasoning (BabyVision: 91%), at 1/3 the cost of M3 and 3x the speed.
API pricing (flat rate): $0.15 / $0.60 per 1M input/output tokens, with cache reads at $0.015 / 1M tokens.
Model weights will arrive on Oct 23.
API: platform.minimax.io
M Plan: minimax.io/m-plan
MiniMax Code for desktop: agent.minimax.io/download
NVIDIA releases Nemotron-3.5-Bolt-E4B, a new model that runs locally on 6GB RAM. š
Bolt-E4B can run on edge devices like phones and delivers top-tier performance for its size. Trained on Gemma 4, it supports vision.
GGUF: huggingface.co/unsloth/NVIDIā¦
Guide: unsloth.ai/docs/models/nemotā¦
Available now: NVIDIA Nemotron 3.5 Bolt E4B 4GB ā”ļø
Our newest small open model is built for local assistants, interactive characters and robots that can carry on a conversation and use tools to get things done.
Watch Reachy Mini chat and move, with Bolt handling conversation and NVIDIA Cosmos handling vision. Both models run locally on Jetson Orin Nano 8GB.
Learn more: research.nvidia.com/labs/nemā¦
Unsloth AI retweeted
We plugged Beam into @opencode, asked it to look at @UnslothAI, and create a fine-tuning notebook for the latest and smallest Gemma-4 model on a Text2SQL task ā an out of distribution domain for Beam.
The model successfully researches the model cards, Unsloth documentation, and available READMEs, and implements the fine-tuning logic and surrounding machinery. As a result, Beam is able to successfully finetune Gemma, increasing Gemmaās accuracy on its heldout test set by 66.5%.
Run Flint locally on 24GB RAM via Unsloth Dynamic GGUFs!
Flint is the strongest MoE model under 50B parameters.
Recursiveās Apache-2.0 model has 39.8B parameters (2.5B active). Run at 125 t/s on Mac, 150 on RTX Spark
GGUF: huggingface.co/unsloth/flintā¦
Guide: unsloth.ai/docs/models/flint
Introducing Flint, @Recursive_SIās first model: a fast local coding model built with our automated research system, the Eureka Machine.
⢠Best-in-class fast local coding
⢠Up to 2x faster than comparable models: 125+ tokens/sec on a MacBook Pro, 150+ on RTX Spark
⢠Built from scratch in under 4 months
Private, unmetered intelligence on your own machine. Open weights available now on @huggingface, run it locally on your favorite framework or try it for free on @opencode or @pidotdev.
š„³ Ornith-1.5-122B Unsloth GGUFs are now available!
š§” Thanks for the awesome work by @UnslothAI
š huggingface.co/unsloth/Ornitā¦
Aloha!šŗ Meet Ornith-1.5-122B: No Tradeoff Between Strong and Fast
Stronger. Faster. Leaner. Ornith-1.5-122B delivers frontier-level capabilities with exceptional inference efficiency. With superior instruction following, stronger task decomposition, and deeper understanding of real-world workflows, it achieves state-of-the-art performance among open-source models of comparable size, while running on a single NVIDIA DGX card!!
Open weights under the MIT license.
Coding
ā
SWE-bench Verified: 0.845
ā
TB2.1: 0.813
ā
CyberGym: 0.800
ā
NL2Repo: 0.512
Agentic
ā
HLE: 0.446
ā
BrowseComp: 0.785
ā
MCP-Atlas: 0.746
ā
Toolathlon: 0.708
š¤Hugging Face: huggingface.co/ornith-ai/Ornā¦
We canāt wait to see what you build with Ornith-1.5-122B.
DeepSeek V4.1-Flash-1017 can now run locally! š³
We made 1-bit retain ~88.2% accuracy while shrinking size from 508GB to 250GB (ā52%).
Run on a 128GB+ RAM device.
V4.1 Flash 1017 is the strongest open model to date
Guide: unsloth.ai/docs/models/deepsā¦
GGUF: huggingface.co/unsloth/DeepSā¦
š DeepSeek-V4.1-Flash-1017 is here!
Stronger benchmark performance, faster task completion, and shorter outputs than the previous versionāa Pareto improvement in performance and cost. Now live on our website, app, and API.
š ļø Optimized for @DeepSeekHarness, Tencent CodeBuddy, WorkBuddy, and more. Upgrade to DeepSeek Harness v0.2.1-rc.1 or later for the best experience with the new model.
1/4
You can now run and train 120B+ parameter models locally on a laptop! NVIDIA RTX Spark is here. š
With Unsloth and RTX Spark, train diffusion, decision, multimodal models ~2.7Ć faster with optimized kernels and binaries for CUDA on ARM. All local with up to 128GB unified memory
NVIDIA RTX Spark is here.
Build AI locally, bring your creative ideas to life, and take your favorite games along.
A new beginning for @Windows PCs starts today.



































